Use std::move on this pointer as the default approach to supporting
rvalue ref-qualifiers if a user really wants to wrap.
std::move requires <memory> headers so add swigfragments.swg for all
languages to use common fragments. Just header file fragments for now.
Internally, handle function ref-qualifiers in the function decl type string.
Needed for a whole host of things to work like %feature and %rename.
Add %feature %rename and %ignore testing for ref-qualifiers.
Fixes#1059
Methods with rvalue ref-qualifiers are ignored by default as it is not
possible to have an rvalue temporary from the target language (which is
needed to call the rvalue ref-qualified method).
A warning 405 is shown mentioning the ignored rvalue ref-qualifier method
which can be seen with the -Wextra option.
cpp_refqualifier.i:15: Warning 405: Method with rvalue ref-qualifier ignored h() const &&.
Usually rvalue and lvalue ref-qualifier overloaded methods are written - the
lvalue method will then be wrapped.
Fixes#1051. Using declarations to templates were missing in SWIG's internal typedef tables.
This led to a few problems, such as, templates that did not instantiate and generated
C++ code that did not compile as SWIG did not know what scope the template was
in. This happened mostly when a using declaration was used on a template type in a
completely unrelated namespace.
Fix some cases of type lookup failure via a combination of both using directives and
using declarations resulting in C++ code that did not compile as the generated type was
not fully qualified for use in the global namespace. Example below:
namespace Space5 {
namespace SubSpace5 {
namespace SubSubSpace5 {
struct F {};
}
}
using namespace SubSpace5;
using SubSubSpace5::F;
void func(SubSubSpace5::F f);
}
The scoping rules around %template have been specified and enforced.
The %template directive for a class template is the equivalent to an
explicit instantiation of a C++ class template. The scope for a valid
%template instantiation is now the same as the scope required for a
valid explicit instantiation of a C++ template. A definition of the
template for the explicit instantiation must be in scope where the
instantiation is declared and must not be enclosed within a different
namespace.
For example, a few %template and explicit instantiations of std::vector
are shown below:
// valid
namespace std {
%template(vin) vector<int>;
template class vector<int>;
}
// valid
using namespace std;
%template(vin) vector<int>;
template class vector<int>;
// valid
using std::vector;
%template(vin) vector<int>;
template class vector<int>;
// ill-formed
namespace unrelated {
using std::vector;
%template(vin) vector<int>;
template class vector<int>;
}
// ill-formed
namespace unrelated {
using namespace std;
%template(vin) vector<int>;
template class vector<int>;
}
// ill-formed
namespace unrelated {
namespace std {
%template(vin) vector<int>;
template class vector<int>;
}
}
// ill-formed
namespace unrelated {
%template(vin) std::vector<int>;
template class std::vector<int>;
}
When the scope is incorrect, an error now occurs such as:
cpp_template_scope.i:34: Error: 'vector' resolves to 'std::vector' and
was incorrectly instantiated in scope 'unrelated' instead of within scope 'std'.
Previously SWIG accepted the ill-formed examples above but this led to
numerous subtle template scope problems especially in the presence of
using declarations and using directives as well as with %feature and %typemap.
Actually, a valid instantiation is one which conforms to the C++03
standard as C++11 made a change to disallow using declarations and
using directives to find a template.
// valid C++03, ill-formed C++11
using std::vector;
template class vector<int>;
Similar fixes for defining classes using forward class references have
also been put in place. For example:
namespace Space1 {
struct A;
}
namespace Space2 {
struct Space1::A {
void x();
}
}
will now error out with:
cpp_class_definition.i:5: Error: 'Space1::A' resolves to 'Space1::A' and
was incorrectly instantiated in scope 'Space2' instead of within scope 'Space1'.
* fflexo-javalist:
Java std::vector minor improvement
Fix Java container tests for change in vector constructor declaration
Add in missing Java std::list listIterator index range checking
Minor correction in C# std::list doNextIndex
Add missing typedefs to Java std::vector
Consistent destructor declarations
Remove Java std::list::max_size
Java std::list std::vector - test addAll and subList
Handle length_error exceptions in Java std::vector::reserve
Remove Java std::list::assign
Additional add/remove methods added to Java std::list wrappers
More efficient add implementation for Java std::list
Java std::vector std::list: add missing exception handling
Java std::vector std::list enhancements
Modify std::list declarations to match the C++ standard
Fix removing elements from std::list Java wrapper
Improve Java std::list std::vector runtime tests and wrap std::list::clear
Wrap std::list::empty as isEmpty in Java
javabase typemap improvement for std::list
Java std::list - fully qualifiy Java class name to avoid potential name ambiguity
cosmetics
Remove redundant code
Java std::list rework to be consistent with std::vector wrappers
li_std_list testcase not working for most languages
re-enabled li_std_list test
Switched from autobox to jboxtype per #842
Document autobox.i
Made the conversion from long->int for size_type mapping onto Java interfaces cleaner.
Be consistent in semantics of %extend on std::list::iterator
Comment on consideration of making iterator non-static.
Java style fix: iterator->Iterator
Moving iterator functionality into nested Java class now.
Removed typedef from li_std_list test as it's not expected to work properly in templated code
Added a best case workaround for std::list::size_type vs jint problem. There's a bit of commentry added around it too for clarity.
Drop non-const reference from autobox typemap macro to be consistent.
just use a forward declaration for C++ iterator types to fix enum errors
Added enum to li_std_list tests
Added li_std_list to the Java test-suit makefile
added more comments in a few places
Base _runme.java for li_std_list off li_std_vector_runme.java
Expose more types from li_std_list.i
Don't expose sort() to avoid adding dependencies on all std::list users
Target each method specificly for setting modifiers
Don't expose remove() method from std::list to avoid confusing it with Java's remove() in List
- added std_list.i implemenatation that extends Java's AbstractSequentialList base class - added autobox.i that provides supporting typemaps for generics in containers
- Add missing vector copy constructor
- Add constructor to initialize the containers. Note that Java's
equivalent constructor for ArrayList just sets the capacity, whereas
the wrappers behave like the C++ constructor and set the size. I've
done this mainly because there has been a vector(size_type) constructor
in the Java wrappers for many years, so best to keep this unchanged.
* vadz-java-vector:
Fix potential STL std::vector wrappers <: digraphs problems.
Add runtime checks for vector size in Java
Make std::vector<> wrappers conform to List interface in Java
Add helper macro to avoid duplication in Java vector typemaps
Conflicts:
CHANGES.current
For shared_ptr proxy proxy classes, add a protected method swigSetCMemOwn for modifying
the swigCMemOwn and swigCMemOwnDerived member variables which are used by various other
methods for controlling memory ownership.
Closes#230Closes#759
- Fix java_director_typemaps test name clash ensuring parallel tests work
- Replace assert with code that is guaranteed to run
- Nicer generated code from directorargout typemap
* fschlimb-templ_def_cache:
Add template_default_cache runtime tests
Fix template_default_cache testcase
template_default_cache is a multi-module test
using 2-level caching as suggested by @wsfulton
account for explicitly qualified scopes
adding test
restricting chaching template default types
Derive the class wrapping std::vector<> in Java from java.util.AbstractList<>.
This makes it possible to use it with various algorithms working with Java
collections and, maybe even more importantly, makes it possible to iterate
over the wrapped vectors using for-each loops.
This commit is based on the original patch by Volker Diels-Grabsch from
https://sourceforge.net/p/swig/patches/278/
- directorargout didn't work - it used string "jresult" instead of
argument name
- java/typemaps.i didn't work with directors
- test for using java/typemaps.i in directors added
